Ncloth Medal String Simulation Progress

In the current project I am working on, Cutting Losses, there is a scene where the character grabs a medal hanging on a wall hanger. The medal is attached to a string, and I applied nCloth to simulate this string.

Old Ugly Joint and CTL Medal Rig

At first, I was planning to create joints, connect them with controls and constraints, and animate them one by one. However, I decided to use nCloth for a more natural and realistic movement.

The medal string is supposed to be a thick, sturdy material that does not stretch too much. To achieve this, I went through several rounds of adjustments. Since I don’t have much experience with nCloth, this process was quite challenging for me. Throughout the process, I relied on help from ChatGPT and searched google multiple times. (No Youtube videos)

The problems I encountered during this simulation process were:

1. The medal string stopped and didn’t sit properly on the metal ring.

2. The metal ring itself moved or wobbled unstably instead of staying fixed.

3. The string slipped off the ring and fell through.

4. When the string touched the wall hanger, instead of smoothly sliding off, it either got stuck or bounced unnaturally.

5. The string stuck to the hanger and stretched unnaturally.

String goes through the ring with no strength

If you look closely at the medal, you’ll see two metal rings connecting the medal and the string. I set up these two rings in a parent-child relationship, with the main geometry as the parent. I applied a passive collider to the ring that connects directly to the medal string. The wall hanger that holds the medal was also set up with a passive collider.

• I added a ring helper collider as an invisible barrier to prevent the string from slipping through the metal ring.

Two Metal Rings Connecting Medal and String

Ring Help Collider's Collisions

To resolve these issues:

1. I used helper (invisible) colliders as ‘invisible walls’ to support the physical limitations of simulation.

2. Collision problems were mostly solved by a combination of smoother geometry, higher collide strength, and higher solver quality.

3. Bouncing issues were resolved by setting bounce=0, increasing collide strength, and adjusting friction to the right level.

4. Overstretching was controlled by increasing stretch resistance and adjusting self-collision settings.

Through all these steps, I was able to solve each simulation problem one by one and finally found the best settings that worked for this scene!

Since these were problems I tackled on my own, the result may not be perfect — but I was able to achieve an outcome I’m fairly satisfied with.

Works Well!

Dynamic Properties of the STring

Collision Value of the string